web-dataviz-d3

Pass

Audited by Gen Agent Trust Hub on Sep 8, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill provides patterns for generating interactive tooltips that interpolate data directly into the DOM using the .html() method without explicit sanitization.
  • Ingestion points: Data properties such as d.label, d.value, and d.properties.name are used to populate tooltip content in examples/interaction.md and examples/advanced.md.
  • Boundary markers: The instructions do not define boundary markers or separators to prevent the agent from interpreting embedded instructions in these data fields.
  • Capability inventory: The skill utilizes d3-selection for direct DOM manipulation and event handling within the SVG/HTML context.
  • Sanitization: No sanitization or escaping mechanisms (such as DOMPurify or conversion to .text()) are implemented for the external data used in HTML tooltips, which could lead to cross-site scripting (XSS) if the visualization data is attacker-controlled.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 8, 2026, 03:01 AM
Security Audit — agent-trust-hub — web-dataviz-d3